AI engineering glossary · ML fundamentals
What is concept drift?
Concept drift: A change over time in the relationship between the inputs and the target itself, so patterns the model learned no longer hold. It requires retraining or adaptation, and is detected by watching live performance.
Concept drift: an example
After a new regulation, the same transaction features no longer signal fraud in the way they did before.
Related terms
More ML fundamentals terms
- Cost-sensitive learning
- Covariate shift
- Cross-validation
- Curse of dimensionality
- Data leakage
- Data snooping
- Data-centric AI
- Dataset
Where you learn it
Concept drift is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.